홈 › Papers

Safety-Regulated Transfer Reinforcement Learning with Adaptive Teacher Guidance

2026-06-25 · Wenjie Huang, Yang Li, Jingjia Teng, Mingwei Jin, Kai Song, Yougang Bian, Yongfu Li, Qisong Yang, Helai Huang arxiv

We propose Safety-Regulated Adaptive Transfer Reinforcement Learning (SRATRL), a teacher--student framework that combines safety-triggered intervention, safety-adaptive value shaping, and policy-compatibility-based optimization for efficient target-domain adaptation. First, a safety-triggered closed-loop intervention strategy is developed that activates teacher guidance according to the instantaneous safety cost and adaptively adjusts the intervention threshold based on the student policy's recent safety performance, thereby providing timely safety supervision while progressively restoring student autonomy as its safety improves. Next, a safety-adaptive teacher-guided value-shaping scheme is introduced, in which a teacher-consistency signal is incorporated into the critic target, and its contribution is dynamically regulated by the safety-constraint multiplier, enabling stronger teacher guidance under elevated safety risks and gradually weakening such guidance as the safety constraint is better satisfied. In addition, a teacher-student policy-compatibility weighting approach is proposed to alleviate the adverse optimization effects caused by policy mismatch. It reweights teacher-intervened transitions according to the relative likelihood of the executed action under the teacher and student policies, thereby improving policy-update stability. Experimental results demonstrate that compared with a Proximal Policy Optimization with Lagrangian constraint baseline, the proposed method improves the average velocity by 6.90%, and reduces the crash ratio by 75.00%. These results demonstrate that the proposed method can reduce safety costs while maintaining competitive task efficiency.

📄 PDF Abstract BibTeX arXiv:2606.26527

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning Transfer Learning Domain Adaptation

Simplex-enabled Safe Continual Learning Machine

2024-09-05 · Hongpeng Cao, Yanbing Mao, Yihao Cai, Lui Sha 외

This paper proposes the SeC-Learning Machine: Simplex-enabled safe continual learning for safety-critical autonomous systems. The SeC-learning machine is built on Simplex logic (that is, ``using simplicity to control com…

Continual Learning Deep Reinforcement Learning

AUKT: Adaptive Uncertainty-Guided Knowledge Transfer with Conformal Prediction

2025-02-23 · Rui Liu, Peng Gao, Yu Shen, Ming Lin 외

Knowledge transfer between teacher and student models has proven effective across various machine learning applications. However, challenges arise when the teacher's predictions are noisy, or the data domain during stude…

Autonomous Driving Conformal Prediction image-classification Image Classification+2

Physics-Regulated Deep Reinforcement Learning: Invariant Embeddings

2023-05-26 · Hongpeng Cao, Yanbing Mao, Lui Sha, Marco Caccamo

This paper proposes the Phy-DRL: a physics-regulated deep reinforcement learning (DRL) framework for safety-critical autonomous systems. The Phy-DRL has three distinguished invariant-embedding designs: i) residual action…

Deep Reinforcement Learning reinforcement-learning Reinforcement Learning

Soft Action Priors: Towards Robust Policy Transfer

2022-09-20 · Matheus Centa, Philippe Preux

Despite success in many challenging problems, reinforcement learning (RL) is still confronted with sample inefficiency, which can be mitigated by introducing prior knowledge to agents. However, many transfer techniques i…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Safer Reinforcement Learning through Transferable Instinct Networks

2021-07-14 · Djordje Grbic, Sebastian Risi

Random exploration is one of the main mechanisms through which reinforcement learning (RL) finds well-performing policies. However, it can lead to undesirable or catastrophic outcomes when learning online in safety-criti…